ChiCTR2600120307 版本V1.0 版本创建时间2026/03/12 08:57:43 中国临床试验注册中心

审核状态:

Project audit state:

通过审核

Successful

注册号:

Registration number:

ChiCTR2600120307 

最近更新日期:

Date of Last Refreshed on:

2026-03-12 08:57:11 

注册时间:

Date of Registration:

2026-03-12 00:00:00 

注册号状态:

补注册

Registration Status:

Retrospective registration

注册题目:

AI辅助医生 vs 医生 vs AI:两阶段临床诊断与管理任务比较研究

Public title:

Physician+AI?vs?Physician?vs?AI:?A?Two-Stage?Comparative?Study on?Diagnostic?and?Management?Tasks

注册题目简写:

English Acronym:

研究课题的正式科学名称:

AI辅助医生 vs 医生 vs AI:两阶段临床诊断与管理任务比较研究

Scientific title:

Physician+AI?vs?Physician?vs?AI:?A?Two-Stage?Comparative?Study on?Diagnostic?and?Management?Tasks

研究课题代号(代码):

Study subject ID:

在二级注册机构或其它机构的注册号:

The registration number of the Partner Registry or other register:

申请注册联系人:

杨晓楠 

研究负责人:

杨晓楠 

Applicant:

Xiaonan Yang 

Study leader:

Xiaonan Yang 

申请注册联系人电话:

Applicant telephone:

+86 10 8896 4614

研究负责人电话:

Study leader's
telephone:

+86 10 8896 4614

申请注册联系人传真 :

Applicant Fax:

研究负责人传真:

Study leader's fax:

申请注册联系人电子邮件:

Applicant E-mail:

15730596690@163.com

研究负责人电子邮件:

Study leader's E-mail:

15730596690@163.com

申请单位网址(自愿提供):

Applicant website(voluntary supply):

研究负责人网址(自愿提供):

Study leader's website(voluntary supply):

申请注册联系人通讯地址:

北京市石景山区八大处路33号

研究负责人通讯地址:

北京市石景山区八大处路33号

Applicant address:

No.33, BaDaChu Road, Shijingshan District, Beijing 100000, China

Study leader's address:

No.33, BaDaChu Road, Shijingshan District, Beijing 100000, China

申请注册联系人邮政编码:

Applicant postcode:

研究负责人邮政编码:

Study leader's postcode:

申请人所在单位:

中国医学科学院整形外科医院

Applicant's institution:

Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

研究负责人所在单位:

中国医学科学院整形外科医院

Affiliation of the Leader:

Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

是否获伦理委员会批准:

Approved by ethic committee:

Yes

伦理委员会批件文号:

Approved No. of ethic committee:

( 2025 )注册第( 315 )号

伦理委员会批件附件:

Approved file of Ethical Committee:

查看附件View

批准本研究的伦理委员会名称:

中国医学科学院整形外科医院医学伦理委员会

Name of the ethic committee:

The?Review?Board?of?Plastic?Surgery?Hospital, Chinese?Academy?of?Medical?Sciences

伦理委员会批准日期:

Date of approved by ethic committee:

2025-10-14 00:00:00

伦理委员会联系人:

裴晔

Contact Name of the ethic committee:

Ye Pei

伦理委员会联系地址:

北京市石景山区八大处路33号

Contact Address of the ethic committee:

No.33, BaDaChu Road, Shijingshan District, Beijing 100000, China

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 10 5396 8034

伦理委员会联系人邮箱:

Contact email of the ethic committee:

研究实施负责(组长)单位:

中国医学科学院整形外科医院

Primary sponsor:

Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

研究实施负责(组长)单位地址:

北京市石景山区八大处路33号

Primary sponsor's address:

No.33, BaDaChu Road, Shijingshan District, Beijing 100000, China

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

北京

市(区县):

石景山

Country:

China

Province:

Beijing

City:

Shijingshan

单位(医院):

中国医学科学院整形外科医院

具体地址:

北京市石景山区八大处路33号

Institution
hospital:

Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

Address:

No.33, BaDaChu Road, Shijingshan District, Beijing 100000, China

经费或物资来源:

中国医学科学院整形医学研究专项

Source(s) of funding:

Plastic?Medicine?Research?Fund?of?Chinese?Academy of?Medical?Sciences

研究疾病:

无  

Target disease:

None

研究疾病代码:

Target disease code:

研究类型:

干预性研究

Study type:

Interventional study

研究所处阶段:

其它 

Study phase:

N/A

研究设计:

随机平行对照 

Study design:

Parallel 

研究目的:

主要目的:验证“医生+AI辅助”较“医生+传统资源(禁用AI)”在诊断与管理综合评分上的优效。 次要目的:比较用时、答案长度/结构、分维度得分、错误/有害建议率与依从度;刻画与“AI单独作答”的相对水平。  

Objectives of Study:

Primary Objective: To verify the superiority of "Physician + AI assistance" over "Physician + traditional resources (AI prohibited)" in the composite score for diagnosis and management. Secondary Objectives: To compare time spent, answer length/structure, dimension-specific subscores, rates of errors/harmful recommendations, and adherence; and to characterize performance relative to "AI-only responses."

药物成份或治疗方案详述:

 

Description for medicine or protocol of treatment in detail:

 

纳入标准:

1.≥18岁;2.执业/注册医师(含住培/主治/副高/正高);3.同意录屏与数据使用;4.可在指定时段完成线上答题。

Inclusion criteria

1.Age >=18; 2. licensed/registered physicians (including residents, attendings, associate-senior, and senior physicians); 3.provide consent for screen recording and data use; 4.able to complete the online assessment within the specified time window.

排除标准:

1.不同意录屏或数据使用;2.无法保证在对照组完全禁用AI;3.存在影响作答的客观障碍(设备/网络/认知等)。

Exclusion criteria:

1. Decline screen recording or data use; 2. unable to ensure complete prohibition of AI in the control group; 3. presence of objective barriers that affect response (e.g., device/network/cognitive issues).

研究实施时间:

Study execute time:

From 2025-10-14 00:00:00 To 2026-03-25 00:00:00  

征募观察对象时间:

Recruiting time:

From 2025-10-29 00:00:00 To 2026-03-25 00:00:00

干预措施:

Interventions:

组别:

医生 + AI辅助

样本量:

100

Group:

Physician + AI assistance

Sample size:

干预措施:

在统一平台按病例分步作答,可使用指定AI工具

干预措施代码:

Intervention:

Completes stepwise case tasks on a unified platform with access to the designated AI tool.

Intervention code:

组别:

医生+传统资源(禁用AI)

样本量:

100

Group:

Physician + traditional resources

Sample size:

干预措施:

在统一平台按病例分步作答,可以查阅传统医学相关资料,不能使用任何AI工具

干预措施代码:

Intervention:

Completes stepwise case tasks on a unified platform. Relevant materials on traditional medicine can be consulted, and no AI tools of any kind may be used.

Intervention code:

研究实施地点:

Countries of recruitment and research settings:

国家:

中国

省(直辖市):

北京 

市(区县):

石景山 

Country:

China

Province:

Beijing

City:

Shijingshan

单位(医院):

中国医学科学院整形外科医院 

单位级别:

三甲 

Institution
hospital:

Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

专家盲评综合评分(百分制)

指标类型:

主要指标

Outcome:

Expert-blinded composite score (0–100)

Type:

Primary indicator

测量时间点:

每位受试医师完成全部病例作答后,集中进行双评盲评与仲裁评分

测量方法:

Measure time point of outcome:

After each participant completes all cases; centralized, blinded double-rating with adjudication

Measure method:

指标中文名:

分维度得分(如:诊断、管理、人文沟通等)

指标类型:

次要指标

Outcome:

Dimension-specific subscores (e.g., diagnosis, management, communication)

Type:

Secondary indicator

测量时间点:

与综合评分同步评定

测量方法:

Measure time point of outcome:

Assessed concurrently with the composite score

Measure method:

指标中文名:

总用时 / 每病例用时

指标类型:

次要指标

Outcome:

Total time / per-case time

Type:

Secondary indicator

测量时间点:

作答期间实时记录;任务结束后汇总

测量方法:

Measure time point of outcome:

Recorded continuously during the task; aggregated post-completion

Measure method:

指标中文名:

答案长度与结构(字数、要点覆盖)

指标类型:

次要指标

Outcome:

Answer length & structure (word count, key-point coverage)

Type:

Secondary indicator

测量时间点:

提交答案时自动/人工提取

测量方法:

Measure time point of outcome:

Extracted at submission

Measure method:

指标中文名:

错误/有害建议率

指标类型:

次要指标

Outcome:

Rate of errors / harmful recommendations

Type:

Secondary indicator

测量时间点:

评分阶段由专家依据预设清单标注

测量方法:

Measure time point of outcome:

During expert scoring using a predefined checklist

Measure method:

指标中文名:

依从度(资源使用符合性;对照组禁用AI的遵循度)

指标类型:

次要指标

Outcome:

Adherence/compliance to resource rules (incl. AI prohibition in control)

Type:

Secondary indicator

测量时间点:

作答期间通过录屏核查;事后质控复核

测量方法:

Measure time point of outcome:

Checked via screen recording during the task; verified in post-hoc QC

Measure method:

指标中文名:

评分者一致性(κ/ICC)

指标类型:

次要指标

Outcome:

Inter-rater reliability (κ/ICC)

Type:

Secondary indicator

测量时间点:

评分完成后计算

测量方法:

Measure time point of outcome:

Calculated after completion of scoring

Measure method:

指标中文名:

安全性与过程事件(数据安全事件、显著心理不适、严重流程偏离等)

指标类型:

次要指标

Outcome:

Safety & process events (data-security incidents, notable distress, major protocol deviations)

Type:

Secondary indicator

测量时间点:

全程监测,事件触发即记录与上报

测量方法:

Measure time point of outcome:

Monitored throughout; event-driven recording and reporting

Measure method:

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

组织:

Sample Name:

None

Tissue:

人体标本去向

其它  

说明

Fate of sample:

0thers  

Note:

征募研究对象情况:

Recruiting status:

正在进行

Recruiting

年龄范围:

Participant age:

最小 Min age 18 years
最大 Max age years

性别:

男女均可

Gender:

Both

随机方法(请说明由何人用什么方法产生随机序列):

分层区组随机(变长区组),计算机生成随机数序列 由独立统计学家(不参与受试者招募与纳入)使用经验证的软件(如 R 或 SAS)按专科与职称进行分层,在各分层内采用1:1 随机区组分配,区组长度设置为 4/6/8 随机变长以降低可预测性;设置并记录固定随机种子(seed)、生成时间与软件版本,形成计算机生成的随机数字序列。随机序列仅由不参与纳入的专人/数据管理员保管,并供中央随机系统调用。

Randomization Procedure (please state who generates the random number sequence and by what method):

Stratified permuted-block randomization (variable block sizes) with computer-generated sequence An independent biostatistician (not involved in participant recruitment or enrollment) will generate a computer-based random number sequence using validated software (e.g., R or SAS). Randomization is 1:1 within strata defined by medical specialty and professional rank, employing permuted blocks with variable sizes (4/6/8) to reduce predictability. A fixed random seed, generation time, and software version will be recorded. The allocation sequence is stored securely by a non-enrolling data manager and implemented via a central randomization system.

是否公开试验完成后的统计结果:

Calculated Results after the Study Completed public access:

公开/Public

盲法:

盲法类型:评分者盲法(assessor blinding);统计分析阶段组别遮蔽(analysis blinding)。受试医师与干预不可盲。 实施对象: 评分专家:对随机组别(A:医生+AI;B:医生+传统资源)及“AI单独作答”的来源信息均不知情。 统计分析人员:在主分析完成与代码冻结之前,仅接触以 A/B 哑变量标识的组别信息。 实施过程: 提交答案后,由与纳入无关的质控人员对文本与日志进行去标识(移除AI使用痕迹、时间戳/用户名、系统提示等可识别信息),并随机化呈现顺序。 评分采用双评+仲裁:两名独立专家在统一rubric下盲评,分歧由第三专家或共识会议仲裁。 评分平台对评分者隐藏元数据(作答时长、资源使用记录等),必要信息仅用于依从性核查且与评分界面物理隔离。 统计分析在锁库后以预先登记的分析方案执行,组别在主要结果产出前保持遮蔽;敏感性/亚组分析遵循同样遮蔽流程。 盲态维持与解盲: 访问控制与审计日志确保评分材料与组别映射分离;发生重大偏离或安全事件需解盲时,按SOP由PI与数据管理员共同审批,记录原因与时间。 研究结束后统一解盲并归档。

Blinding:

Type of blinding: Assessor blinding and analysis blinding. Participants/interventions cannot be blinded. Who is blinded: Expert raters are blinded to randomized assignment (Arm A: Physician+AI; Arm B: Physician+traditional resources) and to whether a response originates from the "AI-only" reference arm. Statisticians/analysts are group-masked (A/B dummy labels) until completion of the primary analysis and code freeze. Procedures: After submission, materials are de-identified by independent QC staff (removing AI cues, timestamps/user IDs, system prompts, etc.) and presented in randomized order. Scoring uses independent dual ratings with adjudication under a standardized rubric; discrepancies are resolved by a third rater or consensus. The scoring interface hides metadata (e.g., time-on-task, resource logs). Compliance data are collected separately and isolated from raters. Statistical analyses follow a pre-registered analysis plan on a locked dataset, with group masking maintained until primary results are finalized; sensitivity/subgroup analyses adhere to the same masking. Maintaining and breaking the blind: Role-based access and audit trails ensure separation between scoring materials and allocation. Unblinding for serious deviations or safety reasons follows SOP with PI and data manager approval, documenting rationale and timing. Global unblinding occurs after study completion and archiving.

试验完成后的统计结果(上传文件):

Calculated Results after
the Study Completed(upload file):

是否共享原始数据:

IPD sharing

是Yes

共享原始数据的方式(说明:请填入公开原始数据日期和方式,如采用网络平台,需填该网络平台名称和网址):

主要研究数据可在论文正式发表后90日内联系通讯作者获取;数据涉及受试者隐私,将保存在本地数据库,如确有需要,可联系通讯作者请求获取。

The way of sharing IPD”(include metadata and protocol, If use web-based public database, please provide the url):

The main research data can be obtained by contacting the corresponding author within 90 days after the official publication of the paper. Since the data involves the privacy of the subjects, it will be stored in a local database. If there is a genuine need, please contact the corresponding author to request access to the data.

数据采集和管理(说明:数据采集和管理由两部分组成,一为病例记录表(Case Record Form, CRF),二为电子采集和管理系统(Electronic Data Capture, EDC),如ResMan即为一种基于互联网的EDC:

本研究数据采集:回顾性部分采用问卷星调查,数据储存在本地及腾讯云备份,确保安全性;前瞻性部分采用腾讯问卷调查,数据储存在本地数据库及腾讯云部分。 本研究数据管理:本研究所有相关数据将储存在研究者本地及腾讯云备份中,确保安全性,为保护受试者隐私,如有公开要求及获取需求,请联系通讯作者。

Data collection and Management (A standard data collection and management system include a CRF and an electronic data capture:

Data Collection in This Study: The retrospective part adopts the Wenjuanxing survey (a professional online survey platform in China), and the data is stored locally with backup on Tencent Cloud to ensure security; the prospective part uses Tencent Questionnaire, and the data is stored in the local database and partially on Tencent Cloud. Data Management in This Study: All relevant data of this study will be stored locally by the researchers with backup on Tencent Cloud to ensure security. To protect the privacy of participants, if there are requirements for data disclosure or requests for data access, please contact the corresponding author.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

暂未确定/Not yet

注册人:

Name of Registration:

 2026-03-12 08:57:11